Colonization and Changing Social Structure:
Kazakhstan, 1896-1910
Gani Aldashev
SBS-EM, ECARES, Université libre de Bruxelles and CRED, University of Namur
Catherine Guirkinger
Department of Economics and CRED, University of Namur
March 2016
ECARES working paper 2016-10
ECARES
ULB - CP 114/04
50, F.D. Roosevelt Ave., B-1050 Brussels BELGIUM
www.ecares.org
Colonization and changing social structure: Kazakhstan, 1896-19101
Gani Aldashev
ECARES, Université libre de Bruxelles (ULB), and CRED, University of Namur
42 Avenue F Roosevelt, 1050 Brussels, Belgium
E-mail: [email protected]
Catherine Guirkinger
Department of Economics and CRED, University of Namur
8 Rempart de la Vierge, 5000 Namur, Belgium
E-mail : [email protected]
1
This paper was first presented at the conference “Role of History and Diversity in Economic
Development” (New Economic School, Moscow, October 4-5, 2015). We thank Zhuldyzbek
Abylkhozhin, Avner Greif, Ekaterina Khaustova, Andrei Markevich, Dilip Mookherjee, Nathan
Nunn, Andy Newman, Jean-Philippe Platteau, Aset Temirgaliev, and the participants of the
Moscow conference and the seminar at Boston University for useful suggestions.
1
Colonization and changing social structure: Kazakhstan, 1896-1910
Abstract
This paper investigates how, under increasing land pressure during Russian settlement
in Kazakh steppes in the late-XIXth century, family-based institutions and social
structure of Kazakhs evolved to adapt to new economic conditions. Using a rich
dataset constructed from Russian colonial expedition materials, we find that during the
transition from nomadic pastoralism to a semi-sedentary pastoralist-agriculture
system, the size of Kazakh extended families increased, those of communes and clans
decreased, and that Kazakhs identified stronger with lower levels of genealogical clan
system. Within families, property rights on land became more individualized,
households became less likely to pool labor to farm, and wage labor contracts in
agriculture became common. We discuss theoretical explanations for the observed
patterns.
1. Introduction
Both development economists and economic historians have recently been debating about
the role families, kinship networks and clan play in shaping individual incentives and
2
determining aggregate economic outcomes and development trajectories. Researchers have
documented the importance of such institutions for migration decisions, occupational choice,
credit, public good provision, and numerous other aspects of economic life (see, for instance,
Leunig et al., 2011; Gupta, 2014; Wegge, 1998; LaFerrara, 2003; and Greif and Tabellini, 2015).2
The importance of these customary institutions have been shown in highly diverse contexts,
spanning all areas of the developing world (for India, see Platteau, 1995, Munshi and
Rosenzweig, 2006, and Gupta, 2014; for China, see Freedman, 1958, and Greif and Tabellini,
2015; for Mexico, see Munshi, 2003; for Sub-Saharan Africa, see Platteau, 2000, and LaFerrara,
2003, etc.).
Institutions, however, are not fixed in time and tend to evolve in response to changes in
the socio-economic environment. Historians and anthropologists have long been concerned with
the evolution of family institutions and the role of economic factors behind such evolution (see,
for instance, Goody, 1983; Todd, 2011). Similarly, the study of institutional change has been for
long time an active area of research in economic history (see, for instance, Davis and North,
1971; North, 1990; Greif, 2006, for classic contributions). Development economists, however,
tend to ignore changes in family institutions when examining the impacts of policies or changes
in resource endowments, making the assumption that these institutions (such as co-residence
patterns, inheritance practices, or marriage patterns) change very slowly, so that taking them as
given is not problematic (see Fafchamps and Quisumbing, 2007, and Cox and Fafchamps, 2007,
for reviews). Recent contributions challenge this assumption and study how household
composition (and, in some cases, pre-mortem inheritance practices) change in response to
2
Excellent surveys of this literature are provided by Cox and Fafchamps (2007), LaFerrara
(2007), and Munshi (2014).
3
technological change (Foster and Rosenzweig, 2002), raising land pressure (Guirkinger and
Platteau, 2015), land policies (Bardhan et al., 2014), or programs of public transfers (Hamoudi
and Thomas, 2014). The latter two studies indicate that ignoring impacts of policy on household
divisions may lead to substantial biases in the evaluation of policies.
One of the difficulties in studying change in family institutions in general is limited data
availability. Ideally, one needs a panel data including both the measures of institutions and
behavior (preferably, at micro-level), with sufficiently large time frame. Moreover, such data
should come from episodes or periods of relatively large-scale changes in the economic
environment of the society under study. Clearly, such a combination of circumstances and data is
rare.
In this paper, we exploit one such dataset that we have constructed using Russian colonial
statistical expedition materials in Central Asia. More specifically, we focus on the extendedfamily and clan-network institutions of Kazakh nomadic pastoralists in the late 19th – early 20th
century. This was a period of massive Russian peasant in-migration into Central Asia, a
migration that implied a sharp increase in land pressure in Northern, Western, and Central
Kazakhstan. We document how the traditional institutions of Kazakhs changed in response to
rising land pressure and new agricultural technologies that Russian colonization brought about. In
particular, we document changes in co-residence patterns (i.e. the size of extended families
sharing the same winter stop and communes sharing summer pastures) and in production
relations within extended families (the individualization of property rights on land and the
development of labor markets). We also explore how the observed changes in family
organization can be understood theoretically as rational responses to the new economic
environment. Our main findings can be summarized as follows. Russian peasant in-migration
4
increased land pressure and contributed to transfer of agricultural technology of sedentary
peasants to Kazakh pastoralists. These factors induced Kazakhs to switch from nomadic
pastoralism to semi-sedentary mixed economy. In the process accompanying this economic
transition, the size of Kazakh groups of kin-related households residing together (the “extended
family”) increased, while household size remained roughly constant. Contrarily, communes and
clan groups became smaller, property rights on most valuable land plots became more
individualized within the extended families and common organization of production at the
extended family level gave way to household-based production. Simultaneously, households
increased their reliance on hired labor (with wages paid largely in cash) for crop and hay
production and part of these labor contracts were likely to be established within extended families
(rich households hiring their poor relatives. Given that the institutions governing land and labor
allocation within families are typically considered as being “slow moving” (Roland, 2004), the
changes we document here are remarkably fast. Our panel of provinces indicate for example that
over a 7-10 years period, about 10% of extended families abandoned collective hay making (at
the level of the extended family) and distributed hay plots to individual households composing
the extended family. The same proportion of families stopped to organize crop production at the
level of the extended family (making it each household responsibility instead) and increased their
reliance on hired labor.
2. Data and historical background
2.1.Data sources
For our main source of data we rely on the quantitative materials collected by the Russian
colonial expeditions into Kazakhstan, in two waves (F.A. Shcherbina, 1898, 1902a, 1902b,
5
1903a, 1903b, 1907, 1908; V.K. Kuznetsov 1909, 1910a, 1910b, 1910c, 1910d; P.P.
Hvorostanskii 1912a, 1912b). The first-wave expedition, headed by Fedor Shcherbina, lasted
from 1896 to 1903 and covered 12 provinces in 3 regions in Western, Northern, and Central
Kazakhstan.3 Virtually all Kazakh households that had their winter stop in these 12 provinces
were included in the survey. Guirkinger and Aldashev (2016) provide the details of the history of
the expedition, its objectives, the type of materials that were collected, as well as a summary of
the analyses by historians of the reliability of the data collected by the expeditions. In brief, this
data constitutes highly reliable and detailed agricultural censuses: The results for each province
were published as a separate volume, each containing a descriptive part and a series of annexes
(including the original variables recorded and aggregated at various levels). The annexes (from
which we extracted most of the information used in our analyses) contain tables with
demographic (age and gender structure), economic (livestock wealth, cultivation, labor relations),
technological (techniques of agriculture, tools used), and institutional (land-use and landownership institutions) variables.
The major goal of the Shcherbina expedition was to calculate the approximate number of
Russian settlers that these provinces could accommodate. The expedition built its conclusions
regarding the amount of land that could be used for Russian peasant settlers on the basis of the
explicit assumption that (most of) Kazakh population would maintain its nomadic economic
lifestyle. However, the General Department was disappointed with the conclusions of the study
3
As in Guirkinger and Aldashev (2016), we adopt the following convention in translating the
names of administrative levels created by the Russian administration in the Kazakh steppes: the
large administrative area (oblast) corresponds to a region, its sub-division (uezd) to a province,
and the smaller administrative area (volost) to a district.
6
(Petrova 1979) and considered that the rapid sedentarization of Kazakhs that was under way
implied that substantially more land could be freed up for peasant settlement. It subsequently
decided to finance another wave of expeditions. This second wave took place between 1907 and
1913 and consisted of 5 separate smaller expeditions. Three expeditions (the one headed by
Kuznetsov and covering the entire Akmolinsk region, the second headed by Perepletchikov and
covering 3 of 5 provinces of the Semipalatinsk region, and the third headed by Hvorostanskii and
covering the entire Turgay region and 3 of 4 provinces of Ural’sk region) allowed to obtain
repeated observations for 10 of 12 provinces analyzed in the first wave. In addition, two
expeditions (Skryplev expedition covering the entire Syr-Darya region and Rumyantsev
expedition covering the entire Semirechie region) collected data from the Southern part of
Kazakhstan that was not reached by the first-wave expedition.
In this paper, we use aggregate data at the province level from the two waves of
expeditions for 7 provinces of Northern and Central Kazakhstan (Aktuybinsk, Kustanay, and the
entire Akmolinsk region with its 5 provinces), as well as extended family level data from the first
wave for the same provinces (approximatively 11300 observations). In addition, for one province
(Petropavl) we were able to match the extended-family questionnaires across the two waves of
expeditions and constructed a panel dataset with 1335 extended families. Figure 1 shows the
geographic location of these provinces.
2.2.Kazakh nomadic-pastoralist economy
Nomadic pastoralism became the dominant economic form in Kazakhstan around 1000 BCE
(essentially because the climatic conditions for agriculture worsened around that period) and
remained so until the middle of the 19th century, when Russian in-migration into the Kazakh
steppes took off. The nomadic pastoralist system relied on herding, and thus livestock represented
7
the main stock of wealth, as well as the key production input and the principal source of food. In
terms of livestock composition, Kazakhs mainly held and bred horses and sheep; once the
shortening of nomadic trajectories started, as described below, cattle also became an important
category of livestock. The nomadic pastoralism in Kazakhstan consisted of seasonal
transhumance, i.e. of changing physical location of households and its livestock between two and
four times during the calendar year. These transhumance between summer and winter pastures
(with relatively shorter stays on intermediate autumn and spring stops) was necessary because
livestock under this system lived throughout the year on natural grass cover as fodder; thus,
remaining permanently on the same place would have rapidly become unsustainable.4
Kazakh nomadic pastoralism was a carefully balanced system developed through
centuries of adaptation to the geography and the climate of the area. Summer pastures provided
abundant fodder during the warmer months; however, these areas became inhabitable in winter.
Distances between the winter and summer pastures were large, often exceeding 200 kilometers
one way (Matskevich 1929; Ferret 2014). The scarcity of good winter pastures (areas close to
rivers, lakes, and hills) implied the need to preserve the fodder of the winter pasture for the next
year. This need, coupled with the relatively flat landscape in most of the Central and Western
Kazakhstan, resulted in long-distance seasonal transhumance of Kazakhs. A contemporary
observer described the logic of this system as follows:
“For livestock to survive on natural pasture through the winter, it has to grow enough fat
during the summer; this is obtained if summer pastures are sufficiently rich in grass… If summer
pastures are constrained, the livestock is unable to accumulate fat, and thus to avoid that it loses
4
Ferret (2014) provides a detailed classification of main forms of nomadic pastoralism in Central
Asia at the end of the 19th century.
8
weight even further, it becomes necessary to move earlier to the [intermediate] autumn
pastures… The lack at any of these three pastures [summer, autumn, and winter] cannot be
compensated even by excellent quality of the other two; for instance, lack of grass at natural
winter pastures, no matter how abundant are the summer ones, will break the equilibrium of the
nomadic economy and lead to the destitution of Kazakhs…” (Dzhanturin 1883: 16-17)
The harsh climatic conditions of the steppe and the reliance of natural grass cover (rather
than producing and stocking fodder) implied that the nomadic pastoralist system was vulnerable
to climatic shocks. Jut, the Kazakh word for a particularly dry summer (during which animals
were unable to accumulate enough fat) followed by a harsher-than-normal winter, implied huge
generalized losses of livestock. For instance, Tolybekov (1971, p. 541-542) reports that
approximately 59 per cent of total livestock was lost during the jut-year winter of 1879/80 in
Irghiz and Turgay provinces. The frequency of such shocks was high: the winters of 1850/51,
1855/56, 1879/80 and 1891/92 were jut years leading to large-scale losses of livestock
(Tolybekov 1971, p. 542).5
Interestingly, this vulnerability to climatic shocks was substantially mitigated by some of
the technological changes in livestock management that Russian migrants brought into the Steppe
during the 19th century. Contemporary accounts attest that hay-making started in the Kazakh
5
The management of climatic risk to livestock is one of the main problems faces by Mongolia
that currently features the society and economic organization most similar to Kazakhs of the
period under study (see Benson, 2011, and World Bank, 2015). The Mongolian word for such
climatic shock, dzud, is clearly of the same origin as the Kazakh jut.
9
steppes around 1840s-1850s. For instance, Daulbayev (1881), describing the economic
organization of the Kazakhs of Kustanay province between 1830 and 1880, writes:
“[Around 1830] they moved regularly during the winter along those rivers from one place
to another, with their livestock and families, seeking forage for their animals, given that no one
among them prepared hay for winter and did not do any cultivation… After [the administrative]
changes [of 1835-1840s], first the Kazakhs living closest to the Russian settlements, and later
also others, taking their Russian neighbors as examples, started to prepare hay for their livestock
for winter and to build winter enclosures for their animals” (Daulbaev 1881: 99, 113).
Several other authors indicate that this technique was learnt by Kazakhs from their Russian
neighbors (Katanaev, 1904; Shcherbina 1908: 202-208; Kurylev 1998: 34-35). Crucially, these
changes implied that the period of the winter stop could be lengthened, as livestock no longer
depended solely on the natural grass cover available at the winter pasture. The positive effect of
this innovation was that the animals could survive even during the jut years.
2.3.Family and clan institutions of Kazakhs
The nomadic-pastoralist society of the Steppe was structured around the clan system,
which consisted of complex networks of blood-related lineages. A strict exogamy rule banned
marriages within the same clan: more precisely, relatives up to the seventh degree could not
marry each other. The clan identity was transmitted from fathers to sons, whereas women
integrated their husband’s clan. This social organization was structured in several layers (see
Figure 2), each with a specific economic and social role.
The smallest unit was a household consisting of a married couple with several children
and, sometimes, other close relatives. A household held private property on livestock but not on
land: the smallest unit with claims on land was the extended family (called aul-q'stau in Kazakh
10
and khozyajstvenniy aul in Russian). This was a small community of several nuclear households
(around 9-10 households in our data) that were typically closely related by kin. The extended
family spent the whole year together, its households living at a short distance from each other
during winter at the winter stop and migrating – together with other extended families of the
same clan – to the summer pasture.
Several extended families jointly constituted a clan (ata-balasy or “descendants of the
same grandfather” in Kazakh), whereas several clans composed a tribe (ru in Kazakh).6 Land on
winter stops was closed-access common property resources of extended families, whereas
summer stops were common property resources at the clan level. Access to pastures was carefully
regulated within the clan.
There was a substantial degree of inequality within extended families, mostly because
livestock ownership was on the basis of nuclear households. This meant that an average extended
family included one or two wealthy households (“bay”) as well as their poorer relatives, which
were often economically dependent on bay’s household. Tolybekov (1971) writes: “From the
outside, [an aul] looked like a certain amount of temporarily built yurts [wooden-structure houses
covered with felt]… There was no common ownership of belongings. Households represented
economic units having livestock as their private property… Wealth inequality existed between
them” (Tolybekov 1971: 500).
6
Tribes themselves entered into one of the three larger confederations or hordes: Senior,
Middle, and Junior (juz in Kazakh). These upper layers of the social structure (ru and juz) played
the role of regulating inter-clan conflicts and managing diplomatic relations with the neighboring
countries.
11
The main role of the extended family (headed by a patriarch called xxxx, which stands for
white beard) was the management of land on winter pastures – the main asset for nomadic
pastoralists in the arid steppes of Central Asia. The economic environment related to land on such
pastures could substantially affect the composition and size of extended families: “The auls of
nomadic pastoralists varied in size, and changed even within a year, which occurred under the
influence of social and natural conditions, such as the wealth of households, quality of grass
cover, etc.” (Tolybekov 1971: 500-501).
The next level of social structure – the clan – fulfilled several other key functions. It
played a central role in nomadic production through its coordination of joint transhumance, to
and from the summer pasture, and the regulation of access to land on the summer pastures.
Towards the end of winter, the heads of extended families belonging to the same clan sent
messengers to each other, to agree on the timing of migration to the summer pasture that they
jointly exploited (Chormanov 1906). The coordinated move to the summer pastures helped
organize the defense against possible raids of livestock thieves during the move, facilitated the
appropriation of summer pastures (as land rights on summer pastures were relatively loosely
defined and inter-clan conflicts over their boundaries were common), and enabled a clan to
exploit economies of scale in caring after the herds (Masanov, 2011: 408).
Second, the clan provided insurance against shocks to its members. For instance, if a
family lost livestock to predators or a particularly harsh winter, other members of the clan
provided the family with some livestock (Vladimirtsov, 1934). The geographic spread between
winter pastures of the members of the same clan often was quite large; thus, in case of a climatic
12
shock (a particularly harsh winter) in one area, the members of the clan having winter stops in
other areas contributed to partially cover the livestock losses of the former.7
2.4.Russian resettlement and colonization
In the first half of the 18th century, facing extended wars with their Eastern neighbors
from China (Oirats), Kazakh tribes officially requested to become a protectorate of the Russian
Empire in the first half of the 18th century. The Kazakhs of the Junior Horde were the first to
request this status in 1731, followed by the Middle and Senior Hordes in 1735 and 1748. Through
the 19th century, the Russian emperors gradually transformed the protectorate status of the
Kazakh Steppes into that of a colony through a series of political and administrative reforms and
military interventions. These reforms started in 1822 with the abolition of the khanate of the
Middle Horde and terminated in 1868 with the declaration that the entire territory of Kazakhstan
was under the control of the Russian Empire (Abuseitova et al., 2001, pp. 353-359).
The migration of Russians into Kazakhstan that started in the 17th century was initially
small but accelerated in to the last quarter of the 19th century, reaching its peak in the 1910s. It
developed in two phases (Cossack military and poor Russian peasant); it was the second and
largest phase that led to fundamental structural changes in the Kazakh nomadic economy. This
phase started in 1861 (Galiev, 2009: 223; Demko 1969: 52), after the abolition of serfdom.
Between 1861 and 1889, the first wave of peasant settlers started to arrive into the Steppe. This
in-migration was limited and somewhat chaotic. Although these peasants migrated without State
encouragement and planning, the Czarist administration tolerated this migration because it eased
7
For a discussion of other roles played by clans, see pp. 88-90 of Guirkinger and Aldashev
(2016).
13
land pressure in the European part of Russia. The year 1889 marked the adoption of the First
Resettlement Bill: the State started to actively encourage peasant in-migration into the Kazakh
Steppes and tried to regulate it. This bill offered the Russian landless peasants land “for free”, in
the amount of 15 desyatinas (approximately 16.4 hectares) per household, in the Asian part of the
Russian Empire (Olcott, 1995: 87). The Resettlement Administration (Pereselencheskoe
upravlenie) was created, and in 1895 the Czarist government organized and financed the
statistical expeditions into the Steppe (we use the materials of these expeditions in this paper).
Finally, after 1906 (the year of the start of Stolypin agrarian reforms), peasant resettlement
became an imperial priority and turned into the fully-fledged colonization, aiming at maximizing
the use of land resources throughout the Russian empire.
As reported by Demko (1969), in 1897 the Russian-speaking population of the four Kazakh
regions directly bordering with Russia (Uralsk, Turgay, Akmolinsk, and Semipalatinsk)
comprised 496 thousand people, corresponding to 20.6 per cent of the total population of these
regions. By 1905 this figure increased to 844 thousand people, corresponding to 28.9 per cent of
the total population.
This massive in-migration both discouraged nomadic pastoralism and encouraged sedentary
agriculture at winter stops (soil and climatic conditions at summer pastures did not allow crop
cultivation). On the one hand, the large-scale occupation of pasture lands and transhumance
routes made nomadic pastoralism more costly, because, as Sedelnikov (1907) noted:
“Reduction in pastures led to an increasing death of livestock in winter... and this forced
weaker and poorer tribes to re-consider their future: given that the previous form of the
economy could not provide their subsistence, they had to look for another one that better
14
corresponds to the new situation... And now these tribes sedentize in the north to live there
for the entire year, close to and partially under the protection of Russian villages” (p. 23).
In addition, there was knowledge transfer concerning crop cultivation and the relevant
agricultural tools. Demko (1969) mentions that the example of settlers influenced numerous poor
Kazakhs, particularly those with relatively small herds, to attempt crop cultivation. Such transfer
occurred to a large extent through direct observation:
“[Some Kazakhs] have been stopping by the Russian towns and observing how the Russian
ploughed. At first they hired Russian peasants for agricultural work, but when they saw
how profitable the agriculture could be, they started to plough themselves. Thus, according
to the testimony of the Kazakhs themselves, they started laboring the soil and even now it is
practiced under the influence and with the direct participation of the Russian population,
[...] mostly peasants” (Tikhonov 1903, p. 69)
Russian colonization also substantially modified the property rights on land. Until 1891, the
land legally belonged to Kazakh tribes. In that year, the Rulings Concerning the Administration
of Akmola, Semipalatinsk, Semirechinsk, Ural, and Turgay Regions stated (Article 119) declared
that the land occupied by nomads was the property of the State (Zimanov 2005: 500-518). This
regulation did grant Kazakh nomads usufruct rights on the land that they occupied for pastures;
however even these rights could be revoked (Article 120). The ruling officially gave the Kazakhs
rights equivalent to those of Russian peasants (Article 11). One should note, however, that the
formal land titles introduced by the Tsarist administration were regularly trampled by colonial
settlers; thus, the equality of rights applied only to Kazakhs who conducted cultivation and only
to those plots that served agriculture (Martin, 2001).
15
As the above descriptions show, the family and social and organization Kazakh nomads
was intimately related to their production system. Each layer of this organization played a welldefined role in the allocation of productive resources (e.g. clans as manager of summer pastures,
whereas extended families as managers of winter pastures). We now proceed to exploring how
clan and family institutions adjusted to the large-scale changes in the production system triggered
by Russian peasant colonization We start by documenting the transformations in the production
system, in particular, in families relatively close to Russian settlements. We then investigate
changes in family and clan institutions per se. More specifically, we focus on three dimensions of
these institutions: co-residence rules at the winter stops and summer pastures, property rights
over land within the extended families, and labor allocation within the families. For each of these
dimensions, we describe changes over time (using our panel of provinces) and the correlation
between the institutions and the geographic proximity to Russian settlements (using the extendedfamily dataset from the first wave). We then proceed to a discussion of the economic mechanisms
driving these changes.
3. Changes in the production system
The massive Russian in-migration triggered deep changes in the production system of
Kazakhs, towards more intensive uses of land. First, introduction and massive diffusion of the
practice of haymaking (consisting in cutting natural grass, drying and storing it to feed livestock
during the cold season) enabled to sustain more animals on a given land area during the winter.
This resulted in longer stays at the winter stops and shorter transhumance distances. Second, crop
cultivation – also implying a more intensive use of land – gradually became an increasingly
important source of supplementary nutrition for Kazakhs.
16
To document these changes, we first present the correlation between the reliance on more
land-intensive techniques and the geographical closeness of Kazakh winter stops to Russian
settlements, using the large cross-section of extended families surveyed by the first expedition.
We then show changes over time in the same variables at the more aggregate level, by using the
panel of provinces.
Around 1900 the concentration of Russian settlements was greatest in areas close to
province capitals. Over time, the density of settlements increased almost everywhere, however it
still remained highest in the vicinity of the province capitals. This can be seen from the
comparison of Panels A and B of Figure 3. We thus expect that both the increase in land pressure
and diffusion of more modern agricultural technologies are stronger in areas closer to the
province capitals, and that the distance from the winter stop to the province capital is negatively
correlated with the adoption of more land-intensive production techniques. Relatedly, in terms of
the timing of adoption, we expect that these transformations must have occurred earlier in such
closer areas
These expectations are confirmed in our data. Table 1 breaks down the sample by decile
of distance to the province capital and reports, for each decile, the share of extended families that
cultivate crops (column 1), the number of years since they started cultivating (column 2), and the
share of extended families that produce hay (column 3). We observe a monotonic decrease in the
fraction of cultivating families, from 0.85 in the first decile to 0.45 in the last one. Similarly, the
experience of cultivation is longer in areas closer to the province capital. Finally, production of
hay is nearly universal in the first 6 deciles, whereas it is present in only about 80% of families in
the last two deciles. Definitions and descriptive statistics for all the variables are presented in
Table 2.
17
Regression analysis confirms these findings. Table 3 reports the results of OLS
regressions at the extended-family level forthe propensity to cultivate (columns 1 and 2) and the
propensity to produce hay (columns 5 and 6). Columns 3 and 4 report the marginal effects of
Tobit regression; the dependent variable is the number of years that the extended family has been
cultivating. The distance to the province capital has a large and significant impact on all the three
variables, even after controlling for the size of the extended family (measured by the number of
nuclear households it contains) and adding province fixed effects. Similar results obtain when we
replace the distance to capital (in verstas) by log distance. The results in columns 2, 4, and 6, for
example, suggest that one standard deviation increase in the distance to the province capital (154
verstas) reduces the propensity of a Kazakh extended family to cultivate by about 15 percentage
points, the average number of years of cultivation by about 7 years, and the propensity to produce
hay by about 15 percentage points.
Turning to the panel of provinces (in Akmolinsk region), the comparison between first
wave and second wave’s production patterns confirms that a transformation of the production
system was clearly under way. Table 4 indicates that in nearly all the provinces the total number
of haystacks produced increased substantially over the ten-year time period between the two
waves, increasing by 60% for the whole region. Crop production (measured by the total area
devoted to cultivation) also increased substantially (area cultivated increased by 54% for the
whole region) and almost in all provinces, with the exception of Kokchetav province. In this
province also (as well as in Atbasar) the share of households cultivating somewhat declined.
Kuznetsov (1909, 1910c) explains that in these two provinces there was a large heterogeneity in
the change of share of households that cultivated, over the ten-year period between the two
waves. In districts where land was more suitable for agriculture, Kazakhs intensified the
cultivation of crops. Contrarily, in areas with poor land suitability, the “early starter” households
18
realized that crop cultivation was not profitable and thus gave up cultivation. On average, these
two tendencies amounted to aggregate decrease in the share of cultivating households.
4. Changes in family institutions
4.1.Change in co-residence patterns
We know turn to the analysis of the changes in family institutions that occurred over the
same period. We start with a description of co-residence patterns, first at the winter stop and then
at the summer pasture. Recall that the households composing extended families share the same
winter stop and migrate jointly to the summer pasture where they meet other families of their
clans and form one large commune. As the production systems adapt to the new environment
(growing land pressure, availability of techniques and tools), the size of extended families
increases while the size of communes decreases. This is visible in our data and confirmed by
contemporary observers.
Starting with the correlation between distance to the province capital and extended family
size, Table 5 reports the average number of households in extended families by decile of distance
to the province capital. It indicates a clear decrease in family size as the distance grows: in the
first decile extended families count on average more than 2 additional households as compared to
the last decile. Contrarily, the average household size is remarkably similar across deciles (see
the last three columns of Table 5). Table 6 reports the result of simple regressions where the
dependent variable is either the size of the family or of the household and the key variable of
interest the distance to the province capital or its logarithm. It suggests (column 1) that a one
standard-deviation increase in the distance to the province capital is associated with a decrease in
the size of the extended family of 1.08 units (therefore about 6.42 individuals). The size of the
19
household also appears to be negatively (and significantly) correlated with the distance to the
province capital, but the size of the coefficient is small: a one standard-deviation increase in the
distance to province capital is associated with a decrease in the average size of the household
equal to barely 0.046 individuals (column 3).
Turning to the temporal changes captured by the panel of provinces, Table 7 reports the
distribution of extended families by size ranges (1 to 5 units, 6 to 10, 11 to 15, 16 to 20, 21 to 30,
41 to 50 and more than 50)8 and by provinces over time. Comparing the results of the first and
the second wave of data suggests a rightward shift in the distribution, whereby the total number
of households in the first (and sometimes second) category strictly decreases over time while the
number of households in the larger size category increases over time. For instance, in Petropavl
province, an average extended family in 1901 contained 8.6 households and 50.4 individuals,
whereas 7 years later, these measures went up to 9.2 households and 53.9 individuals. This
rightward shift in the size distribution of families can also be seen on Figure 4, which depicts the
1901 and 1908 cumulative distributions for this province. Notice instead, on Figure 5, that there
is no change in the size of nuclear households, suggesting that the observed change in extended
family size is not a demographic (i.e. fertility-driven) one.
The data from Petropavl province also allows us to investigate the mechanism behind the
increase in the size of extended families. Specifically for this province, we can match extended
families in the data from the two waves of the expeditions.9 Table 8 reports the result of this
matching. Column (1) and (2) presents the break-down of the first year sample into a) extended
8
These categories are the ones recorded in the expedition originals in the aggregate data.
9
The volumes of the second expedition report for each extended family the corresponding family
ID in the volume of the first expedition.
20
families that neither marge nor split across waves; b) extended families that merge; c) extended
families that split; d) extended families that are not found in the first wave sample. Columns (4)
and (5) reports the break-down of the second year sample into the same categories a), b) and c)
and into e) families that are only in the second wave sample and f) families for which we could
not find the counterpart in the first wave. The table indicates that the majority of families falls in
the first category (68% of the first year sample) and that there are relatively few merges (7% of
the first year sample). Furthermore, while extended families that have disappeared in the second
wave are slightly smaller, this selection effect cannot account for the overall increase in family
size over time. In fact the average number of households in an extended family in the first wave
remains very similar if we ignore this category (it increases from 8.6 to 8.7 units). Thus the
increase in the size of extended families (that we find on average and in the aggregate) does not
appear to be the result of large-scale merging of extended families or of a selection process
whereby smaller families are the ones that migrate out. The most plausible explanation for the
increase in the number of households composing an extended family must then be that family
splits become less frequent.
Turning to the size of communes, an opposite pattern emerges: the size of communes
decreases over time. Table 9 reports the average number of extended families by commune in
each province of Akmolinsk region for the two years of data. In all provinces the average
commune size decreases substantially and on average the decrease represents 1.7 extended
families (10.1 individuals on average) in less than 10 years. This decrease is driven by an
increase of the number of communes in each province (increase of 13% to 47% depending on the
province).
21
Communes typically consist of several extended families from the same clan. In line with
the decrease in the size of communes, we also find that the size of clans decreased substantially.
Table 10 compares the clan size distribution between the two waves of expeditions, for 5
provinces of Akmolinsk region. We see that the total number of clans increased substantially
across the two waves: in the second wave, depending on the province, 40 to 90 per cent more
clans are enumerated by expedition interviewers. Looking across the size distribution, we observe
that small clans (containing only 1 extended family) increase dramatically as a share of total
clans, whereas large clans (federating 5 or more extended families) command a much small
share. While there is substantial heterogeneity across provinces in the levels, the trends are quite
similar. How can clan identity have changed so drastically over a decade? To understand this
dynamics, it is crucial to remember that Kazakh clans are not disconnected units, but rather
elements of a vast interconnected genealogical tree. Figure 6 presents an example of a section of
such tree for one of the largest Kazakh tribes (Argyn). It is plausible to conjecture that when
asked about his clan, an extended family head declares the branch/level of the genealogical tree
with which the extended family members identify most strongly. If so, one explanation for the
above pattern is that over the decade under study the extended families became more likely to
identify with more recent ancestors in their clan genealogical trees. This conjecture is confirmed
by the head of second-wave expedition in Akmolinsk region; he writes: “The clan system is
getting substantially weaker and a huge mass of auls [extended families] separate away from old
clans into a separate, independent existence… The fall of the importance of the clan system that
affects the restructuring of the economic and daily-life organization of Kazakhs is, in turn, clearly
determined by various influences of the new living conditions [i.e. sedentarization] of Kazakh
population” (Kuznetsov 1910: 56).
4.2.Change in property rights over land
22
The statistical expeditions collected detailed data regarding the level of individualization
of land holdings within the extended family. Specifically, our dataset possesses information on
the rules regulating access of individual households (within the extended family) to plots used for
haymaking. Three main types of rules were used. First, hay plots sometimes were the property of
the extended family as a whole, and all households composing the family jointly exploited this
land. At the other extreme, in some other families, the hay making plots were owned by
individual households. Finally, in certain extended families, households were allocated individual
hay plots on yearly basis (every year the extended family re-allocated plots among its member
households). Sometimes, within the same extended family, a subset of households adopted one
rule (e.g. individual ownership of hay plots), whereas another subset exploited their hay plots
jointly.10
Table 11 reports the relative frequency of extended families with individualized hay plots
and extended families with collective plots by deciles of distance to the provincial capital.
Individualized rights are more prevalent close to the provincial capital, with prevalence rate
above 65% in the first two deciles and lower than 45% in the last two deciles. Table 12 confirms
this correlation in a regression framework, controlling for province fixed effects and the size of
the extended family. Column 1, for instance, indicates that a one standard-deviation increase in
the distance to the provincial capital (154 verstas) is associated with a decrease of 12 percentage
points in the prevalence of individualized hay plots.
10
There was substantial diversity in the way in which the reallocation of hay plots occurred.
Enumerators describe cases where the plots were allocated by a lottery, as well as cases when the
extended family head had a final say on the allocation (Shcherbina 1902a: IX-X).
23
Table 13 presents the statistics on the organization of haymaking in the six provinces for
which we have comparable data. Two facts emerge. First, there is a lot of heterogeneity in
haymaking institutions across provinces: some rely massively on fully individualized plot
allocation between households, whereas in other provinces the common-ownership forms seem to
dominate. Second, for three provinces for which we have data in both waves, there is a clear
tendency towards the reduction of the common-ownership form and an increase in
individualization of haymaking production. Thus, the above increase in family size is
accompanied by an individualization of property rights within families.
This trends towards individualized production of hay is also confirmed by contemporary
observes’ accounts. Kuznetsov (1910) writes: “Joint production of hay can be considered as a
more perfect form of common use, under which one precludes the randomness in the allocation
of quantity and quality of hay associated with the yearly re-allocation arrangement… [In several
cases] we have registered the transition from common use of hay-making plots to the individual
form in the last 7-8 years. The opposite cases of switching from the individual to the common-use
form are extremely rare… Currently, one observes the strengthening of the tendency of some auls
and households to better consolidate the ownership of hay-making plots. These Kazakhs say:
“The absence of boundaries is bad: lots of arguments and fights emerge during re-allocations,
thus it is better that each one has his own plot’…” (Kuznetsov 1910: 114-115)
4.3.Changes in family labor allocation patterns
Another dimension of family institutions we can investigate is how the labor of individual
households is allocated across collective tasks at the level of the extended families and individual
tasks. In particular our dataset contains information about whether households jointly cultivated
crops or not and about their participation to the labor market. The institution of joint cultivation is
24
referred to as supryaga and is different from the collective ownership of hay plots introduced
above. In the latter case, the information explicitly refers to individual / collective ownership of
land, while in the former case, the information concerns labor allocation. The two are clearly
related, as collective hay plots implied that households worked on them jointly (as described in
the published volumes of the expeditions). The reverse, however, is not necessarily true and in
fact, the description of the joint crop cultivation suggests that, at least in some case, crop fields
were the individual property of households.
Regarding the participation in the labor market, our dataset contains information on
whether households within the extended families hired workers for various tasks and whether
some household members worked outside their household. In this category we cannot distinguish
between wage employment (for agriculture and livestock) and other activities (small businesses).
Table 14 reports the average of these variables for each decile of distance to the provincial
capital. Joint crop cultivation (column 1) is more common in the last deciles than in the first ones
(although the increase is not monotonous throughout): in the first decile households joined their
efforts to cultivate crops in 54% of extended families, while in the last decile this figure stood at
86%. Labor hiring for crop cultivation (column 2) follows an opposite trend: it was more
common in families closer to the capital (50% in the first decile) than in those further away (36%
in the last decile). Labor hiring for haymaking was also more common in the first deciles than in
the last ones (although the decrease is neither very steep nor monotonous everywhere; see
column 3). In the case of husbandry, our dataset contains information about whether some
households of the extended families were hiring workers on a yearly basis. This variable is
reported in column 4 and shows again a greater level of development of the labor market closer to
the province capitals and a clear gradient as we move away: in the first decile 50% of extended
families include households who rely on these contracts, while in the last decile this figure is
25
38%. Finally, the supply of labor follows a similar trend with 70% of extended families including
members who work for a wage (or have a small business) in the first decile and 58% in the last
decile (column 5). Comparing the figures reported in columns 2 to 4 to those in column 5 shows
that some extended families both demanded and supplied labor (for instance, summing the share
of families hiring for haymaking to the share of families supplying labor exceeds 100%).
Table 15 confirms in a regression framework that labor markets were more active closer
to the provincial capitals while the joint cultivation of crops by households of the same extended
families was more widespread further away from the capital (controlling for province fixed
effects and family size). Thus, the prevalence of joint crop cultivation in extended families
increase by 5 percentage point on average when the distance to the province capital increases by
one standard deviation (column 1), while the prevalence of hiring for cropping simultaneously
decreases by 17 percentage points.
For changes over time, in the case of joint cultivation and labor market outcomes the
information we have is less systematic than for the other dimensions we analyze in the previous
sections. Specifically we have consistent information across time only for two provinces
(Kustanay and Aktyubinsk). Table 16 presents the statistics on joint cultivation and the
development of labor contract institutions in the period under study, for the two provinces, at the
household level.11 The first column indicates that the share of households engaged in joint
cultivation with other households of their extended family decrease over time from 61% to 42%
in Kustanay and from 57% to 50% in Aktyubinsk. Simultaneously, the share of households hiring
11
Although the unit of observation is the extended family, the dataset contains information on the
number of households within the extended family that use the various types of labor contracts.
We could thus construct the descriptive statistics at the household level.
26
labor for cultivation increase in Kustanay from 16% to 26% (the information is not available for
Aktyubinsk in the second wave). Hiring labor for haymaking also became more prevalent in
Kustanay province (and did not change over time in Aktyubinsk). Finally, the share of
households hiring labor for husbandry activities decreased in Kustanay and increased in
Aktyubinsk but, interestingly, the share of long-term contracts grows: among households hiring
workers, the share of those offering yearly contracts increases from 48 to 84% in the first
province and from 40 to 73% in the other province (and the average number of yearly employees
also increased substantially). Finally, the share of households with members working for a wage
or having a small business increased from 31 to 47% in one province and from 32 to 37% in the
other province.
An analysis of labor market participation at the extended family level (using the second
wave of data) suggests that some labor contracts may have been established within the extended
family. Indeed, 72% of extended families include both households who supply hired labor and
households who hired workers (not shown in the table). Typically, richer households employ
individuals from poorer households to work on the farms of the former. Figure 7 indicates a clear
wealth gradient in the participation to agricultural labor market: rich households (as measured by
livestock ownership) tend to hire workers while poorer households tend to supply workers. As
extended families usually contained both poor and rich households, it is not surprising that they
consist of both labor-supplying and employing units. However, wealth heterogeneity within
extended families was clearly larger than that across extended families, as Figure 8 illustrates.
This figure compares the distribution of household by wealth categories (measured in equivalent
horses) to the distribution of average family wealth (in equivalent horses by household) and
indicates that the former distribution is substantially wider than the latter one.
27
Importantly, although the questionnaire did not record whether the workers hired
originate from the same extended family, a closer look at labor market participation at the
extended family level indicates that although a part of the labor contracts were likely established
between households of the same extended families, some labor contracts must have stretched
beyond extended family boundaries. In fact, in the two Western provinces (Aktyubinsk and
Kustanay) 18% of extended families include only households who hire workers and 10% include
only households who supply workers. Furthermore, in about 25% of extended families, the
number of workers hired exceeds the number of men who are either working for a wage or are
having an independent small business. This implies that more than 25% of extended families did
(also) employ workers from other extended families.
What kind of labor contracts were these? Table 17 presents the examples of contract
details, on the basis of descriptions collected by the expedition members. Two interesting facts
emerge. First, the contracts implied a substantial part of the compensation paid in cash (Russian
rubles). The in-kind part of the payment mostly involved working clothes provided to the worker,
but sometimes also dairy and meat products. Second, the compensation of labor mostly employed
in agriculture and livestock herding was definitely lower than that of wholesale traders, but
higher than that of shuttle traders. Thus, labor hired in agriculture was not necessarily the lowest
paid occupation, which argues against the idea (often proposed in Soviet literature) of severe
exploitation of hired workers by rich livestock owners.
5. Discussion
Our findings described above document important changes in social structures in the
Kazakh population during Russian colonization whereby the size of extended families increased,
28
while clans and communes became smaller and fewer extended families shared common summer
pastures over time. Individual households within extended families enjoyed increasingly
individualized user rights over land (in particular for hay making) and became less likely to
engage in collective crop cultivation with other households of the same extended family.
Some of these observed changes may be the “mechanical” consequences of the increase
in land pressure following the multiplication of Russian peasant settlements: the creation of new
winter stops through splitting of extended families is severely constrained and the pressure on
summer pastures precludes the gathering of multiple extended families. Other changes, notably in
property rights and the labor market, are better interpreted as the endogenous response of the
traditional institutions governing production to the changes in the economic environment. In fact,
we find no indication in writings of contemporaries and historians that the individualization of
property rights and of production within extended family was actively promoted by Russian
authorities.
Economic theory proposes several mechanisms suggesting that the observed
individualization of production may be a direct response to the increase in land pressure and the
development of agriculture following the expansion of Russian peasant settlements. First, as the
size of the group (here, extended family) rises, problems of free-riding in collective production
worsen, raising the relative benefits of having well-defined and individualized property rights
(Chamberlin, 1974; Platteau, 2000; Guirkinger and Platteau, 2015). Second, as the technology
becomes more labor- and care-intensive, the above-mentioned inefficiencies associated with
collective production may more costly. Thus under a technology with sufficiently simple labor
tasks, the monitoring that limits the under-provision of effort is relatively cheap. Once the labor
tasks start to become more complex or the cost of workers’ effort increases (which probably was
29
the case under the transition from extensive herding to sedentary agriculture), the cost of
monitoring rises. This again increases the scope of free-riding in collective production, and thus
the benefit of individualization grows (Boserup 1965; Binswanger and Rosenzweig, 1986).
Recent literature in development economics analyzes the determinants of
individualization of agricultural production within the family (through family splits or
distribution of individual plots to specific members). Foster and Rosenzweig (2002) develop a
model to account for the increase in family-farm splits following the Green Revolution in India.
In their collective-household model, technological change exacerbates intra-family heterogeneity
in productivity leading to stronger disagreements over consumption allocations. Ceteris paribus,
this increases the propensity of a family farm to split into smaller units, leading to more
individualized form of land use. Bardhan et al. (2014) and Guirkinger and Platteau (2015) focus
instead on the free-riding problem in family production. Guirkinger and Platteau (2015) argue
that land scarcity leads to the individualization of family farms, as it exacerbates the cost of
inefficiency associated with free-riding of individual family members. Bardhan et al. (2014)
study the effects of land reforms on family-farms divisions. In contrast with Foster and
Rosenzweig (2002), they predict that technological change would reduce the tendency for freeriding in joint production and, therefore, the rate of family-farm divisions. They do not account,
however, for the possibility that technical progress would also raise the importance of the quality
of labor, as described above, and thereby increase the benefit of more individualized forms of
land-use
In our context, the decreasing importance of collective production is accompanied by an
increased reliance on hired labor. In fact, the picture emerging from the description of the labor
market is that of a very active one. The contrast with the contemporary situation in rural areas of
30
Russia is striking. Chayanov (1925) reports that more than 90% of farm households in pre-1917
Russia relied exclusively on family labor (i.e. household labor) for farming and neither hired nor
supplied laborers for working on other peasants’ farms.12 This difference is even more puzzling,
given that in the nomadic Kazakh system, households’ access to land resources was based on
collective holdings and thus, intuitively, more egalitarian.13 We can only speculate about the
reasons behind this remarkable difference between Russian and Kazakh peasants, but the preexistence of labor exchange in the form of patron-client institutions in the traditional nomadic
economy may have contributed to the rapid development of agricultural labor market. One such
institution in Kazakhstan was called sau’n and consisted of rich households giving/leasing their
livestock (for a given period of time) to poor households inside the extended family. The
members of the poor households had to take care of the livestock leased, and could in exchange
12
Interestingly, in his work, Chayanov (1925) used the so-called zemstvo statistics that were
based on surveys that F.A. Shcherbina contributed to designing.
13
Notably, Chayanov reports very large differences in labor-to-land ratios across Russian
household farms.
31
consume its dairy products.).14 15 This institution is equivalent to labor-exchange institutions
among African pastoralists as reported, for example, by Hill (1970).
Platteau (1995)’s investigation of the transformations of the patron-client relationships in
Asia in the 20th century suggests that, in several contexts, the existence of outside opportunities
for clients contributed to dissolving the patronage institution and its replacement by hired labor.
The increase in the bargaining power of the client forces the patron to change their labor
strategies, possibly towards shorter and monetized labor contracts. An agricultural wage labor
market then develops. In our case, the Russian settlers may have provided these new outside
opportunities. In some instances they have been reported to hire Kazakh workers for example
(Sedelnikov, 1907). Interestingly, Platteau (1995) reports that the new labor contracts were more
14
Historians have mixed view on the exact role that sau’n played: some consider it to be
essentially motivated by solidarity considerations, whereas others (mostly of historians with
Marxist views) see it as a mechanism for creating economic dependence of the poor and thus
political control by the class of rich households. Markov (1976) writes: “Sometimes shepherds
and workers were simply hired, with payment in kind or in cash. But sometimes these shepherds
and workers were the poor relatives of the well-to-do livestock owner, and they migrated and
lived together. In that case the relationships were usually veiled by kinship links. Giving
livestock for pasture [i.e. the institution of sau’n] could take many forms: from the genuine
solidarity and mutually beneficial agreement to some of the harshest exploitation of the person
receiving the livestock” (Markov 1976: 301-302).
15
Our data indicates that the share of households that took livestock for sau’n was already
relatively small around the time of the first-wave expedition. In fact, even among the poorest
households, at most around 10 per cent used sau’n.
32
likely to take the form of “regular farm labor contracts” (similar to the yearly contracts observed
here) in the technologically advanced Indian villages of the “Green revolution belt”. This also
provides an alternative channel for the individualization process whereby poorer households with
increased outside opportunities may negotiate an individualization of the extended family
landholding and may be ready to work for a wage for the richer households of the extended
family. This mechanism is actually present in the model of Guirkinger and Platteau (2014, 2015),
where the patriarch of the family is forced to individualize the farm when household members
have increased outside opportunities.
6. Conclusion
Large-scale colonial settlements often implied sharp changes in the constraints faced by
indigenous populations. On the basis of evidence from the late 19th – early 20th century
Kazakhstan, in earlier papers (Aldashev and Guirkinger, 2012; Guirkinger and Aldashev, 2015))
we have studied the effect of increased resource scarcity during the Russian settlement in
Kazakhstan on excess female mortality and the resulting gender bias in the Kazakh population,
and on the role played by traditional clan-based institutions in determining the adaptation to such
changes. In this paper, we focus on the changes in the traditional social structure and institutions
of Kazakhs, document the extent and speed of such changes, and explore theoretical explanations
behind them.
Our study complements the work of economic historians on Czarist Russia. Most of the
studies (for example, Nafziger, 2010; Dennison, 2011; Chernina, Dower, and Markevich, 2014;
Markevich and Zhuravskaya, 2015) focus on the institutional changes in the early 20th-century
33
Russia. Despite the fact that Russia was one of the largest colonial empires, scarce attention has
been paid to the economic history of Russian colonization (one exception is Natkhov, 2015).
Our findings have interesting implications concerning our understanding of the role of
social structure and “slow-moving institutions” (Roland, 2004) in developing countries. The
informal traditional institutions have usually been considered as changing quite slowly: in fact,
most papers in development economics (with the exception of Comola and Prina, 2014) take the
network structure as given. Our study suggests that the speed and magnitude of change of such
institutions can be quite high, which clearly calls for more in-depth studies of such institutional
change in other contexts.
References
Abuseitova, Meruert, Zhuldyzbek Abylkhozhin, Sergei Klyashtornyj, Nurbulat Masanov, Tursun
Sultanov, and Anatoly Khazanov. Istoriya Kazahstana i Central’noi Azii (The History of
Kazakhstan and Central Asia). Almaty: Daik-Press, 2001.
Aldashev, Gani, and Catherine Guirkinger. “Deadly Anchor: Gender Bias under Russian
Colonization of Kazakhstan.” Explorations in Economic History 49, no. 4 (2012): 399-422.
Bardhan, Pranab, Michael Luca, Dilip Mookherjee, and Francisco Pino. "Evolution of land
distribution in West Bengal 1967–2004: Role of land reform and demographic changes." Journal
of Development Economics 110, no. 1 (2014): 171-190.
34
Benson, Charlotte. Dzud Disaster Financing and Response in Mongolia. Washington, DC: World
Bank Publications, 2011.
Binswanger, Hans, and Mark Rosenzweig. "Behavioural and material determinants of production
relations in agriculture." Journal of Development Studies 22, no. 3 (1986): 503-539.
Boserup, Esther. The Conditions of Agricultural Growth: the Economics of Agrarian Change
under Population Pressure. London: Allan and Urwin, 1965.
Chamberlin, John. “Provision of Collective Goods as a Function of Group Size.” American
Political Science Review 68, no. 2 (1974): 707-716.
Chayanov, Alexander V. The Theory of Peasant Economy. Homewood, IL: RD Irwin, 1925.
Chernina, Eugenia, Paul Dower, and Andrei Markevich. “Property Rights, Land Liquidity and
Internal Migration.” Journal of Development Economics 110, no. 1 (2014): 191–215.
Chormanov, Musa. “Zapiska o Kirgizah Pavlodarskogo uezda (Note on Kirghizs of Pavlodar
Province).” Zapiski Zapadno-Sibirskogo Otdela Imperatorskogo Russkogo Geograficheskogo
Obshchestva (Proceedings of the Western Siberian Section of the Imperial Russian Geographic
Society) 32 (1906): 1–30.
Comola, Margherita, and Silvia Prina. “Do interventions change the network? A dynamic peer
effect model accounting for network changes.” Working Paper, Paris School of Economics, Paris,
35
France, 2014.
Cox, Donald, and Marcel Fafchamps. “Extended Family and Kinship Networks: Economic
Insights and Evolutionary Directions.” In: Handbook of Development Economics, vol. 4,
Amsterdam, North-Holland: Elsevier, 2007, pp. 3711-3784.
Daulbayev, Beybit. “Rasskaz o Zhizni Kirgiz Nikolaevskogo Uezda Turgajskoj Oblasti (The
Description of the Life of Kirghizs (Kazakhs) of Nikolaev (Kustanay) Province of Turgay
Region.” Zapiski Orenburgskogo Otdela Imperatorskogo Russkogo Geograficheskogo
Obshchestva (Notes of the Orenburg Section of the Imperial Russian Geographic Society) 4
(1881): 98-117.
Davis, Lance, and Douglass North. Institutional Change and American Economic Growth.
Cambridge, UK: Cambridge University Press, 1971.
Demko, George. The Russian Colonization of Kazakhstan, 1896-1916. Bloomington: Indiana
University Press, 1969.
Dennison, Tracy. The Institutional Framework of Russian Serfdom. Cambridge, UK: Cambridge
University Press, 2011.
Dzhanturin, Selimgirei. Ocherki Kirgizskogo Konevodstva (Essays on Kirgiz [Kazakh] Horsebreeding). Orenburg, 1883.
36
Fafchamps, Marcel, and Agnes Quisumbing. Household Formation and Marriage Markets in
Rural Areas.” In: Handbook of Development Economics, vol. 4, Amsterdam, North-Holland:
Elsevier, 2007, pp. 3187-3247.
Ferret, Carole. “Discontinuités spatiales et pastoralisme nomade en Asie intérieure au tournant
des XIXe et XXe siècles.” Annales. Histoire, Sciences Sociales 69, no. 4 (2014): 957–996.
Foster, Andrew, and Mark Rosenzweig. "Household division and rural economic growth."
Review of Economic Studies 69, no. 4 (2002): 839-869.
Freedman, Maurice. Lineage Organization in Southeastern China. London: Athlone Press, 1965.
Galiev, Vil, ed. Istoriya Kolonizacii Kazahstana v 20-60-h Godah XIX Veka (The History of
Colonization of Kazakhstan in 1820s-1860s). Almaty, Kazakhstan: Mektep, 2009.
Guirkinger, Catherine, and Jean-Philippe Platteau. "The effect of land scarcity on farm structure:
empirical evidence from Mali." Economic Development and Cultural Change 62, no. 2 (2014):
195-238.
Guirkinger, Catherine, and Jean-Philippe Platteau. "Transformation of the family farm under
rising land pressure: A theoretical essay." Journal of Comparative Economics 43, no. 1 (2015):
112-137.
Goody, Jack. The Development of the Family and Marriage in Europe. Cambridge, UK:
37
Cambridge University Press, 1983.
Greif, Avner. Institutions and the Path to the Modern Economy: Lessons from Medieval Trade.
Cambridge, UK: Cambridge University Press, 2006.
Greif, Avner, and Guido Tabellini. “The Clan and the Corporation: Sustaining Cooperation in
China and Europe.” Working Paper, Innocenzo Gasparini Institute for Economic Research,
Bocconi University, Milan, Italy, 2015.
Gupta, Bishnupriya. “Discrimination or Social Networks? Industrial Investment in Colonial
India.” Journal of Economic History 74, no. 1 (2014): 141–168.
Guirkinger, Catherine, and Gani Aldashev. “Clans and Ploughs: Traditional Institutions and
Production Decisions of Kazakhs under Russian Colonial Settlement.” Journal of Economic
History 76, no. 1 (2016): 76-108.
Hamoudi, Amar, and Duncan Thomas. "Endogenous co-residence and program incidence: South
Africa's Old Age Pension." Journal of Development Economics 109, no. 1 (2014): 30-37.
Hill, Polly. Studies in Rural Capitalism in West Africa. Cambridge, UK: Cambridge University
Press, 1970.
Hvorostanskii, P.P., ed. Materialy po kirgizskomu zemlepol’zovaniju, sobrannye i razraborannye
statisticheskoi partiei Turgajsko-Ural’skogo pereselencheskogo raiona. Tom 4. Kustanaiskij
38
Uezd (Materials on Kirgiz [Kazakh] Land Use, Collected and Developed by the Statistical Party
of Turgay-Ural’sk Resettlement Region. Volume 4. Kustanay Uezd). Orenburg, GUZiZ, 1912a.
Hvorostanskii, P.P., ed. Materialy po kirgizskomu zemlepol’zovaniju, sobrannye i razraborannye
statisticheskoi partiei Turgajsko-Ural’skogo pereselencheskogo raiona. Tom 5. Aktuybinskij
Uezd (Materials on Kirgiz [Kazakh] Land Use, Collected and Developed by the Statistical Party
of Turgay-Ural’sk Resettlement Region. Volume 5. Aktuybinsk Uezd). Orenburg, GUZiZ, 1912b.
Katanaev, Georgiy. Kirgizskij Vopros v Sibirskom Kazach'em Vojske (The Kirghiz Question in
the Siberian Cossack Army). Omsk: Tipografija Okruzhnogo Shtaba, 1904.
Kurylev, V.P. Skot, Zemlya, Obshchina u Kochevyh i Polukochevyh Kazahov (Livestock, Land,
and Commune of Nomadic and Semi-nomadic Kazakhs). St.-Petersburg: Ros.Akad.Nauk, 1998.
Kuznetsov, V.K. Kirgizskoe Hozyajstvo v Akmolinskoi Oblasti. Tom 1. Kokchetavskij Uezd.
Povtornoe Issledovanie 1907 Goda (Kirghiz [Kazakh] Economy in Akmolinsk Region. Volume 1.
Kokchetav Province. Repeated Study of 1907). Saint-Petersburg, GUZiZ, 1909.
Kuznetsov, V.K. Kirgizskoe Hozyajstvo v Akmolinskoi Oblasti. Tom 2. Omskij Uezd. Povtornoe
Issledovanie 1908 Goda (Kirghiz [Kazakh] Economy in Akmolinsk Region. Volume 2. Omsk
Province. Repeated Study of 1908). Saint-Petersburg, GUZiZ, 1910a.
Kuznetsov, V.K. Kirgizskoe Hozyajstvo v Akmolinskoi Oblasti. Tom 3. Petropavlovskij Uezd.
Povtornoe Issledovanie 1908 Goda (Kirghiz [Kazakh] Economy in Akmolinsk Region. Volume 3.
39
Petropavl Province. Repeated Study of 1908). Saint-Petersburg, GUZiZ, 1910b.
Kuznetsov, V.K. Kirgizskoe Hozyajstvo v Akmolinskoi Oblasti. Tom 4. Atbasarskij Uezd.
Povtornoe Issledovanie 1909 Goda (Kirghiz [Kazakh] Economy in Akmolinsk Region. Volume 4.
Atbasar Province. Repeated Study of 1909). Saint-Petersburg, GUZiZ, 1910c.
Kuznetsov, V.K. Kirgizskoe Hozyajstvo v Akmolinskoi Oblasti. Tom 5. Akmolinskij Uezd.
Povtornoe Issledovanie 1909 Goda (Kirghiz [Kazakh] Economy in Akmolinsk Region. Volume 5.
Akmolinsk Province. Repeated Study of 1909). Saint-Petersburg, GUZiZ, 1910d.
LaFerrara, Eliana. “Kin Groups and Reciprocity: A Model of Credit Transactions in Ghana.”
American Economic Review 93, no. 5 (2003): 1730-1751.
LaFerrara, Eliana. “Family and Kinship Ties in Development: An Economist’s Perspective.” In:
Platteau, Jean-Philippe, and Robert Peccoud, eds. Culture, Institutions, and Development: New
Insights into an Old Debate. London, Routledge, 2011.
Leunig, Timothy, Chris Minns, and Patrick Wallis. “Networks in the Premodern Economy: The
Market for London Apprenticeships, 1600-1749.” Journal of Economic History 71, no. 2 (2011):
413–443.
Markevich, Andrei, and Ekaterina Zhuravskaya. “Economic Effects of the Abolition of Serfdom:
Evidence from the Russian Empire.” Working Paper, Paris School of Economics, Paris, France,
2015.
40
Markov, Gennady. Kochevniki Azii. Struktura Khozyajstva i Obshchestvennoj Organizatsii (The
Nomads of Asia: the Economic Structure and Social Organization). Moscow: Izd.-vo
Moskovskogo Universiteta, 1976.
Martin, Virginia. “Nomadic Land Claims in the Colonized Kazakh Steppe.” Difference and
Integration 1, no. 2 (2001): 65–74.
Masanov, Nurbulat. Kochevaja Tsivilizatsija Kazahov (The Nomadic Civilization of Kazakhs), 2nd
edition, Almaty: Print-S, 2011.
Matskevich, Nikolai. “Sravnitel'naya Dlina Kochevok Kazakskogo Naseleniya Byvshej
Semipalatinskoj Gubernii (The Relative Length of Transhumance of Kazakh Population of exSemipalatinsk Governorship).” Zapiski Semipalatinskogo Otdela Obshchestva Izucheniya
Kazakstana (Proceedings of Semipalatinsk section of the Society for the Study of Kazakhstan) 1,
no. 18, (1929): 1–33.
Munshi, Kaivan. “Networks in the Modern Economy: Mexican Migrants in the US Labor
Market.” Quarterly Journal of Economics 118, no. 2 (2003): 549-599.
Munshi, Kaivan. “Community Networks and the Process of Development.” Journal of Economic
Perspectives 28, no. 4 (2014): 49–76.
Munshi, Kaivan, and Mark Rosenzweig. “Traditional Institutions Meet the Modern World: Caste,
41
Gender, and Schooling Choice in a Globalizing Economy.” American Economic Review 96, no. 4
(2006): 1225-1252.
Nafziger, Steven. “Peasant Communes and Factor Markets in Late Nineteenth-Century Russia.”
Explorations in Economic History 47, no. 4 (2010): 381–402.
Natkhov, Timur. “Colonization and Development: The Long-term Effect of Russian Settlement
in the North Caucasus, 1890s–2000s.” Journal of Comparative Economics 43, no. 1 (2015): 76–
97.
North, Douglass. Institutions, Institutional Change and Economic Performance. Cambridge, UK:
Cambridge University Press, 1990.
Olcott, Martha. The Kazakhs. Stanford: Hoover Institution Press, 1995.
Petrova, Tatiana. “Organizacija statisticheskogo obsledovanija kazahskih hozyajstv ekspediciej
F.A. Shcherbiny (The organization of statistical analysis of Kazakh households by F.A.
Shcherbina expedition).” Problemy Istorii SSSR (Problems of USSR History), vol. 9: 75-89.
Platteau, Jean-Philippe. “A framework for the analysis of evolving patron-client ties in agrarian
economies.” World Development 23, no. 5 (1995): 767-786.
Platteau, Jean-Philippe. Institutions, Social Norms, and Economic Development. Amsterdam:
Harwood Academic Publishers, 2000.
42
Roland, Gerard. “Understanding Institutional Change: Fast-moving and Slow-moving
Institutions.” Studies in Comparative International Development 38, no. 4 (2004): 109–131.
Sedelnikov, Timofey. Borba za Zemlyu v Kazahskoi Stepi (The Struggle for Land in the Kazakh
Steppe. Saint-Petersburg: Delo, 1907.
Shcherbina, Fedor, ed. Materialy po Kirgizskomu Zemlepol'zovaniju, Sobrannye i
Razrabotannye Ekspediciej po Issledovaniju Stepnyh Oblastej. Tom 1. Kokchetavskij Uezd (The
Materials on Kirghiz Land Use, Collected and Developed by the Expedition for the Analysis of
Steppe Regions. Vol. 1. Kokchetav Province. Voronezh, Russia: GuZIZ, 1898.
Shcherbina, Fedor, ed. Materialy po Kirgizskomu Zemlepol'zovaniju, Sobrannye i
Razrabotannye Ekspediciej po Issledovaniju Stepnyh Oblastej. Tom 2. Atbasarskij Uezd (The
Materials on Kirghiz Land Use, Collected and Developed by the Expedition for the Analysis of
Steppe Regions. Vol. 2. Atbasar Province. Voronezh, Russia: GuZIZ, 1902a.
Shcherbina, Fedor, ed. Materialy po Kirgizskomu Zemlepol'zovaniju, Sobrannye i
Razrabotannye Ekspediciej po Issledovaniju Stepnyh Oblastej. Tom 11. Omskij Uezd (The
Materials on Kirghiz Land Use, Collected and Developed by the Expedition for the Analysis of
Steppe Regions. Vol. 11. Omsk Province. Omsk, Russia: GuZIZ, 1902b.
Shcherbina, Fedor, ed. Materialy po Kirgizskomu Zemlepol'zovaniju, Sobrannye i
Razrabotannye Ekspediciej po Issledovaniju Stepnyh Oblastej. Tom 5. Kustanayskij Uezd (The
43
Materials on Kirghiz Land Use, Collected and Developed by the Expedition for the Analysis of
Steppe Regions. Vol. 5. Kustanay Province. Voronezh, Russia: GuZIZ, 1903a.
Shcherbina, Fedor, ed. Materialy po Kirgizskomu Zemlepol'zovaniju, Sobrannye i
Razrabotannye Ekspediciej po Issledovaniju Stepnyh Oblastej. Tom 7. Aktyubinskii Uezd (The
Materials on Kirghiz Land Use, Collected and Developed by the Expedition for the Analysis of
Steppe Regions. Vol. 7. Aktyubinsk Province. Voronezh, Russia: GuZIZ, 1903b.
Shcherbina, Fedor, ed. Materialy po Kirgizskomu Zemlepol'zovaniju, Sobrannye i
Razrabotannye Ekspediciej po Issledovaniju Stepnyh Oblastej. Tom 3. Akmolinskij Uezd (The
Materials on Kirghiz Land Use, Collected and Developed by the Expedition for the Analysis of
Steppe Regions. Vol. 3. Akmola Province. Saint-Petersburg, Russia: GuZIZ, 1907.
Shcherbina, Fedor, ed. Materialy po Kirgizskomu Zemlepol'zovaniju, Sobrannye i
Razrabotannye Ekspediciej po Issledovaniju Stepnyh Oblastej. Tom 12. Petropavlovskij Uezd
(The Materials on Kirghiz Land Use, Collected and Developed by the Expedition for the Analysis
of Steppe Regions. Vol. 12. Petropavl Province. Chernigov, Russia: GuZIZ, 1908.
Tikhonov, T.I. “O Nuzhdah Osedlogo i Kochevogo Naselenija Akmolinskoi Oblasti (On the
Needs of Sedentary and Nomadic Population of Akmolinsk Region.” Stepnoj Kraj (Steppe
Frontier) 57 (1903): 19–81.
Todd, Emmanuel. L’origine des systèmes familiaux, Tome 1. Paris: Gallimard, 2011.
44
Tolybekov, Sergali. Kochevoe Obshchestvo Kazahov v XVII - Nachale XX Veka: PolitikoEkonomicheskii Analiz (The Nomadic Society of Kazakhs from the XVII to the early XX
Centuries: A Political-Economic Analysis), Almaty: Nauka, 1971.
Vladimirtsov, B.J. Obshchestvennyj Stroj Mongolov. Mongol’skij Kochevoj Feodalism (The
Social Structure of Mongols: the Mongolian Nomadic Feudalism). Moscow: Izd. AN SSSR,
1934.
Wegge, Simone. “Chain Migration and Information Networks: Evidence from NineteenthCentury Hesse-Cassel.” Journal of Economic History 58, no. 4 (1998): 957–986.
World Bank. Mongolia: Agricultural Sectoral Risk Assessment. Washington, DC: World Bank
Publications, 2015.
Zimanov, Salyq, ed. Qazaqtyn Ata Zangdary (The Ancient World of Kazakh Law), vol. 6.
Almaty: Zheti Zhargy, 2005.
45
Table 1: Production system at the extended family level, by distance to the province capital
Average
Decile of
distance to
province capital
% of
extended
families
cultivating
# years since
started
cultivating (if
cultivate=1)
% of extended
families making
hay
2
76%
17
100%
13
99%
1
3
4
5
6
7
8
9
10
Total
85%
70%
60%
63%
59%
56%
55%
47%
45%
62%
18
15
11
15
11
11
11
11
14
98%
100%
99%
99%
96%
96%
81%
80%
95%
Note: The variable # years since started cultivating is only available for two provinces.
46
Table 2: Descriptive statistics for the sample of extended families
Variable
cultivate
# years
cultivating
hay
distance
ln(distance)
# units
household
size
individual
hay
collective hay
joint
cultivation
hired crop
hired hay
yearly
contract
husbandry
working out
or business
Definition
1 if extended family cultivate
crops
number of years since the
extended family started
cultivating
1 if extended family make hay
distance to province capital in
versta (1.5 km)
Log of distance
number of nuclear units in
extended family
average size of households
composing the extended family
1 if at least some households of
extended family own individual
hay plots – defined when hay=1
1 if (some) households of same
extended family have impartible
ownership over hay plots –
defined when hay=1
1 if at least some households
jointly cultivate crops – defined
when cultivate=1
1 if labor hired for crop
cultivation by at least one
household in extended family –
defined when cultivate=1
1 if labor hired for hay making by
at least one household in
extended family – defined when
hay=1
1 if labor hired on yearly basis for
husbandry by at least one
household in extended family
1 if at least one member of
extended family working out for
wage or has a business
Mean
0.61
Std.
Dev.
0.49
Min
0
13.13
0.94
177.55
0.24
154.03
0
0
1
900
11337
11097
4.72
5.94
1.40
1.47
-2.30
0
6.80
22.5
11097
10168
0.53
0.50
0
1
8565
0.47
0.50
0
1
8722
0.71
0.45
0
1
6315
0.38
0.49
0
1
5534
0.49
0.50
0
1
10621
0.37
0.48
0
1
9696
0.66
0.47
0
1
11270
1.47
0
300
N
11217
6.61
5.94
0
Max
1
22.5
6015
10168
Note: The variable # years since started cultivating is only available for two provinces. The difference in the number
of observations across variables is due to the fact that some pages or portion of pages in the archives were not
readable.
47
Table 3: Distance to province capital and adoption of more land-intensive techniques (at extended-family
level, with province fixed effects)
distance
Ln(distance)
(1)
(2)
OLS
OLS
cultivate
-0.001***
(-33.77)
# units
0.012***
N^
9861
(17.27)
cultivate
-0.058***
(-18.64)
(3)
(4)
# years
cultivating
# years
cultivating
Tobit
-0.045***
(-29.47)
0.015***
0.238***
9861
5747
(20.52)
(11.10)
Tobit
-1.367***
(-14.55)
(5)
(6)
OLS
OLS
hay
-0.001***
(-63.93)
0.412***
0.000
5747
9978
(18.31)
(0.91)
hay
-0.016***
(-10.38)
0.003***
(8.23)
9978
Notes: t-statistics in parentheses, * p<0.10 ** p<0.05 *** p<0.01. For tobit, reported coefficients
correspond to marginal effects at mean. The variable # years since started cultivating is only available
for two provinces. The difference in the number of observations for the number of extended families
plowing and making hay is due to the fact that some pages in the archives were not readable.
Table 4: Changes in production systems (5 provinces of Akmolinsk region, 1896-1909)
Province
Akmolinsk
Atbasar
Kokchetav
Omsk
Petropavl
Year
1896-1900
1909
1897
1909
1896
1907
1901
1908
1901
1908
Haystacks
produced,
mln
2.56
2.65
1.20
1.70
1.46
1.87
0.98
2.14
0.93
3.26
Number
of HHs
that
cultivate
7192
8821
3492
3785
2782
3027
206
2338
2688
5812
Share of
all HHs
that
cultivate
0.53
0.54
0.67
0.55
0.22
0.19
0.03
0.27
0.24
0.44
Total area
cultivated
for crops,
thousand
desyatinas
19.55
22.55
5.79
7.07
5.59
4.69
0.53
5.43
5.91
15.42
48
Table 5: Extended family size, by distance to the province capital
Decile of
distance
1
2
3
4
5
6
7
8
9
10
Total
# households in extended family
mean
8.7
8.7
8.0
7.0
7.5
7.6
7.3
6.7
6.8
6.3
7.5
# members per household
standard
standard
deviation median mean deviation median
6.9
7
5.9
1.4
5.8
6.1
6
6.0
1.4
5.9
6.5
5.2
5.8
5.9
6.7
5.2
5.5
5.0
6.0
7
6.1
6
1.5
6.0
6
1.4
5.8
6
1.6
5.9
6
1.6
5.9
5
1.4
6.0
5
1.5
5.9
5
1.4
5.9
6
1.6
6.0
1.5
5.9
5.8
5.7
5.7
5.8
5.8
5.8
5.7
5.8
Table 6: Extended family size and distance to the province capital
distance
Ln(distance)
N
(1)
# units
-0.0066***
(-14.94)
9990
(2)
# units
-0.2538***
(-5.82)
9990
(3)
unit size
-0.0003**
(-2.27)
9925
(4)
unit size
-0.0222*
(-1.95)
9925
Note: t-statistics in parentheses, * p<0.10 ** p<0.05 *** p<0.01.
49
Table 7: Changes in size distribution of extended families (5 provinces of Akmolinsk region, 1896-1909)
Province
Year
1-5
6-10
11-15
16-20
21-30
31-40
41-50
>50
419
645
336
124
69
18
2
4
142
61
Akmolinsk 1896-1900
421
Atbasar
208
Kokchetav
Omsk
Petropavl
1909
1897
1909
1896
1907
1901
1908
1901
1908
Extended families with __ household units
178
694
607
274
218
457
343
675
231
280
617
637
224
262
495
512
277
114
214
87
34
64
307
121
156
78
108
219
285
69
64
112
52
17
38
43
11
3
1
9
55
19
52
16
78
15
20
48
5
18
2
1
2
4
5
1
3
0
8
Total
0
1525
1
609
1
1617
703
1
1646
0
701
5
0
3
4
1756
785
1304
1357
50
Table 8: Matching of extended family in Petropavl province over the two waves of the panel
1901
1908
(1)
(2)
(4)
(5)
(3)
(6)
Extended families that:
Freq. Percent Average units Freq. Percent Average units
Did not split nor merged
907
67.94
907
65.58
8.5
9.9
Split between 1901 and 1908
188
14.08
396
28.63
10.7
8.6
Merged between 1901 and 1908
89
6.67
43
3.11
14.9
6.3
Existed only in 1901 data
151
11.31
7.7
Appeared in 1908 data
24
1.74
7.7
Unable to match
13
0.94
10.8
Total
1335
100
1383
100
9.7
8.6
Table 9: Change in the size of communes (in provinces of Akmolinsk region)
Province
Year
Number of
communes
Extended families
per commune
Akmolinsk
1896-1900
1909
1897
1909
1896
1907
1901
1908
1901
1908
308
443
122
162
336
679
151
228
270
397
5.0
3.7
5.0
4.3
4.9
2.6
5.4
3.6
4.9
3.4
Atbasar
Kokchetav
Omsk
Petropavl
51
Table 10: Changes in clan identification (5 provinces of Akmolinsk region, 1896-1909)
Uezd
Total
With 1
extended
family
127
Akmolinsk
Freq.
417
Atbasar
Freq.
122
Kokchetav
Freq.
354
Omsk
Freq.
108
Petropavl
Freq.
201
Share
Share
Share
Share
Share
Federating
from 2 to 5
extended
families
212
Clans in wave 1
Federating
more than 5
extended
families
0.20
0.47
0.37
0.17
0.38
0.42
0.20
0.39
0.45
0.16
101
558
40
198
82
386
0.46
0.29
0.20
0.43
0.37
0.18
0.41
0.41
82
0.38
169
0.25
37
0.41
45
0.37
46
0.10
300
0.37
22
0.40
589
0.26
163
0.51
78
0.19
90
Federating
more than 5
extended
families
Total
0.51
45
Federating
from 2 to 5
extended
families
With 1
extended
family
0.30
32
Clans in wave 2
70
261
75
151
233
65
204
83
174
56
34
93
40
61
52
Table 11: Property rights over hay plots within extended families, by distance to the provincial capital
decile of
distance
Individual hay
At least some
collective rights
(including yearly
reallocation)
2
66%
36%
1
65%
3
54%
4
50%
53%
5
50%
52%
6
41%
43%
7
56%
46%
8
53%
52%
9
48%
46%
10
50%
48%
Total
35%
51%
53%
47%
Table 12: Correlation between land and labor allocation and distance to province capital (at extendedfamily level, with province fixed effects)
distance
Ln(distance)
(1)
(2)
(3)
individual
hay
individual
hay
collective hay
+ yearly divide
-0.0008***
(-13.72)
# units
0.0012
N
7404
(1.24)
-0.0355***
(-6.98)
0.0005***
(9.32)
0.0028***
-0.0002
7404
7558
(2.86)
(-0.24)
t-statistics in parentheses * p<0.10 ** p<0.05 *** p<0.01
(4)
collective
hay + yearly
divide
0.0300***
(5.91)
-0.0011
(-1.18)
7558
53
Table 13: Changes in property rights over hay plots in extended families (6 provinces, 1896-1909)
Province
Year
Individual
hay
Yearly
reallocation
Collective
hay
Akmolinsk
1896-1900
1909
1897
1909
0.49
0.55
0.34
0.45
0.04
0.11
0.06
0.14
0.30
0.23
0.55
0.28
Atbasar
Kokchetav
Petropavl
Aktuybinsk
Kustanay
1907
1901
1908
1898-1899
1898
0.53
0.37
0.40
0.74
0.51
0.43
0.06
0.15
0.29
0.14
0.60
0.12
0.16
Mixed (at least
two forms
coexist in same
extended family
0.07
0.11
0.01
0.13
0.18
0.02
0.07
0.17
Notes: The categories are defined only for families making hay. The shares do not sum up to one in the first year as
for some families making hay the variables are not defined (3% of the cases). In most of these cases (82%) it is the
first year that hay is prepared so that the institutional arrangement governing hay plot allocation may not be
defined yet.
54
Table 14: Collective plowing and participation to the labor market, by distance to the provincial capital
(1)
(2)
(3)
(4)
(among
families that
cultivate)
hired labor for
crop (among
families that
cultivate)
hired labor for
hay (among
families that
make hay)
hired labor for
husbandry
(yearly contract)
71%
41%
54%
43%
joint
cultivation
1
2
3
4
5
6
7
8
9
10
Total
54%
79%
73%
66%
81%
71%
72%
74%
86%
71%
59%
31%
31%
32%
37%
44%
29%
36%
36%
39%
58%
50%
43%
47%
45%
47%
45%
48%
51%
49%
(5)
working
out for
wage or
business
50%
70%
42%
76%
33%
36%
37%
35%
30%
32%
38%
38%
74%
68%
67%
65%
60%
63%
62%
58%
67%
55
Table 15: Correlation between labor allocation and distance to province capital (at extended-family level, with province fixed effects)
(1)
distance
Ln(distance)
(2)
(3)
Joint
Joint
hired labor
cultivation cultivation crop
0.0003***
(4.35)
-0.0011***
0.0308***
(-16.84)
(5.86)
# units
0.0125*** 0.0123*** 0.0117***
N
0.0003***
(13.26)
(13.17)
(12.10)
-0.0011***
(4)
(5)
hired labor
crop
hired labor
hay
-0.0688***
(-12.88)
0.0129***
(13.28)
-0.0004***
(-7.47)
0.0195***
(22.87)
-0.0004***
(6)
(7)
yearly
hired labor contract
hay
husbandry
-0.0002***
-0.0088**
(-2.30)
0.0203***
(23.95)
(-4.64)
0.0153***
(18.73)
-0.0002***
(8)
yearly
contract
husbandry
-0.0185***
(-4.67)
0.0155***
(19.29)
(9)
working
out wage
or business
-0.0003***
(-10.16)
0.0230***
(31.69)
-0.0003***
(10)
working
out wage
or business
-0.0088***
(-2.79)
0.0239***
(33.18)
t-statistics in parentheses
* p<0.10
** p<0.05
*** p<0.01"
56
Table 16: Change in joint cultivation and labor market for Kustanay and Aktyubinsk provinces
among HH
cultivating:
province
Kustanay
Aktyubinsk
wave
1
2
1
2
Joint
cultivation
61%
42%
57%
50%
hiring
crop
16%
26%
23%
among
HH
making
hay:
share of
household:
hiring hay
13%
26%
27%
26%
hiring
husbandry
29%
24%
42%
36%
among HH hiring for
husbandry :
yearly
contract
husbandry
48%
84%
40%
73%
# workers
on yearly
contract
1.004
1.805
1.396
1.627
share of
household:
working
for wage
or w/
business
31%
47%
32%
37%
57
Table 17:Examples of labor contracts and self-employment (Petropavl province, 1901)
Type of contract
Annual
Annual
Annual
Seasonal (6 months)
Occupation
Agricultural worker
(batrak)
Total salary
Part in cash
Agricultural worker
(batrak)
49 rubles
19 rubles
Haymaking and
livestock breeding
Agricultural worker
(batrak)
Seasonal (30-40
days)
Mower
Annual or seasonal
(winter)
Shepherd
Seasonal
Woodcutter
Shuttle trader
Wholesale trader
Labor contracts
60 rubles
Part in kind
1 shirt, 1 pair of
trousers, 1 pair of boots
Clothes worth 30 rubles
30 rubles
Clothes
56 rubles
50 rubles
Clothes worth 6 rubles
19 rubles (on
average)
15-20
rubles
66 rubles (on
average, annual)
36 rubles
20 rubles (on
average)
Self-employed
20 rubles
Clothes, depending on
in cash part; plus some
livestock or food
products
Clothes worth 30 rubles;
plus abundant food
products
None
33 rubles (on
average, per
season)
200 rubles
58
Figure 1. Administrative structure of Kazakhstan at the end of the 19th century
Source: Ferret (2014), Map 2.
59
Figure 2. Social structure and pre-colonial property rights in the Kazakh society in the 19th
century
Source: Guirkinger and Aldashev (2016), on the basis of Chapter 7 of Tolybekov (1971).
60
Panel A. Pattern of Russian settlements in Kazakhstan around 1900
Panel B. Pattern of Russian settlements in Kazakhstan around 1915
Figure 3. Evolution of Russian settlements in Kazakhstan in the early 20th century
Source: Demko (1969).
61
1
.8
1
.6
.8
.6
.4
.4
.2
.2
0
0
0
50
100
size
distmb01
150
200
distmb08
Cumulative distribution of number of people in
the extended family, 1901-1908
0
10
20
size
disthh01
30
40
disthh08
Cumulative distribution of number of HH units
in the extended family, 1901-1908
Figure 4. Cumulative distributions of the size of extended families, Petropavl province, 19011908
Figure 5. Stability in the size distribution of household units, Petropavl province, 1901-1908
62
Figure 6. An example of a clan genealogical tree
63
Figure 7. Household labor market participation, Aktyubinsk and Kustanay provinces, 1908
Figure 8. Household wealth distribution, Aktyubinsk and Kustanay provinces, 1908
64